In Agent-Based Modeling for Archaeology: Simulating the Complexity of Societies, authors Iza Romanowska, Colin D. Wren, and Stefani A. Crabtree have produced the first agent-based modeling (ABM) textbook designed for researchers studying the human past and educators interested in teaching computational simulation of the past. As an ABM practitioner in archaeology who learned at a time when general ABM textbooks were unavailable, I welcome the addition of a comprehensive, practical, and coherent manual for the instruction and application of ABM in archaeology.

The authors define ABM as “a type of computer simulation that enables investigation of complex phenomena from the bottom up” (p. 6) through the investigation of the behavior of individuals or agents based on user-defined rules. Practically, for archaeologists who wish to become practitioners of ABM, one must become at ease with reading and writing computer code. Agent-Based Modeling for Archaeology relies on NetLogo, the ABM software program most used by archaeologists. Readers will find themselves immersed in NetLogo computational simulation examples, fully coded in print and online, based on published models from archaeology. The authors provide sufficient background and tools so that a beginner can reproduce common archaeological models in minutes. Moreover, the book is organized coherently and produced slickly, with a beautiful cover, a useful table of contents, and quality graphics and printed text.

The authors organized the book into three parts. Part I consists of Chapters 1–3 and introduces the reader to NetLogo, the basics of computational modeling, and coding basics. Part II contains Chapters 4–6, addressing computational algorithms behind agent behaviors and model building. Part III, or Chapters 7–9, describes, with detailed code, methods for combining ABM with spatial and relational (network) data and the generation of artificial data for model testing and validation. Throughout the book, the authors draw from published archaeological ABM examples surrounding the three human behavioral themes of movement, exchange, and subsistence. The concluding chapters provide best-practices guidance for ABM, a handy glossary, and an appendix listing all models covered in the book and a guide for producing color-blind models.

The authors hope to normalize the use of ABM in archaeology by lowering the most substantial barrier to entry for most social scientists—coding. Collecting and publishing code relevant to archaeology in one place, such as this textbook, is immensely practical in this sense. However, the authors’ reliance on NetLogo requires them to expend text throughout the body of the book on the programming vagaries specific to NetLogo. At times, the book can read like a NetLogo manual rather than a volume about the art and science of the application of ABM in archaeology. As noted by the authors, NetLogo is among several software packages available for ABM. Agent-Based Modeling for Archaeology is a timely work right now, but as developers update NetLogo, hopefully the sole reliance on popular software of the day will not consign this book to the back bin in a few years. The authors chose NetLogo because of its ubiquity in archaeological ABM practice and low barriers to entry, but other considerations are important—namely, broad scientific usability, robust algorithmic development and advancement, and the long-term availability of expert and peer support.

Overall, the book will be most useful to two groups of people: instructors of computational social simulation and the beginning learner. Instructors will find code needed for practical lessons as well as suggested exercises at the end of chapter. For the beginner practitioner, this textbook is the necessary “all-in-one” manual for ABM and archaeological application. Most archaeologists should appreciate Chapters 3 through 6—arguably the strongest portions of the work—in which an array of behavioral algorithms describing movement (e.g., random walks and targeted walks), exchange (e.g., price setting and content bias), and subsistence (e.g., Lotka-Volterra and patch choice) are modeled through coding in ABM. However, new theoretical and methodological developments in ABM and archaeology are out of the scope of this book. For example, Chapter 8 relies on NetLogo's underdeveloped algorithms for relational or networked data, resulting in the weakest chapter in the work. Chapter 4 mentions machine learning (ML) briefly in relation to the use of ML for model selection. The incorporation of both network science and ML into ABM is transforming computational social simulation, and although deep treatment of these topics may be out of scope of the current textbook, the field looks forward to a possible second volume discussing advanced methods in ABM and archaeology.